NVIDIA HGX H200 8-GPU Server

Name: NVIDIA HGX H200 8-GPU Server
Brand: NVIDIA
SKU: 935-24287-2740-000

MPN935-24287-2740-000

NVIDIA HGX H200 is the eight-GPU Hopper platform with 141 GB of HBM3e per GPU and NVLink at 900 GB/s. The air-cooled 8-GPU baseboard carries part number 935-24287-2740-000 and ships inside OEM servers.

Source to orderConfirmed per quote

Price on written offer

Send quantity and destination. We confirm supply, lead time and a fixed price in writing.

Units required

Request a quote

Call sales +971 58 561 7897 Email this item

Exact part number 935-24287-2740-000. No edition substitution without your written approval.
Stated per offer: manufacturer, OEM or supplier backed.
Lead time confirmed in your written offer. Agreed per order (Incoterms 2020).
30% deposit to secure the allocation, balance before dispatch. Bank transfer against pro forma invoice, or card payment link for approved orders. Card details are entered on Network International's secure payment page, never on our servers.
Lead-time orders are secured with a 30% deposit, balance before dispatch. Written offers are valid for the period stated, usually 48 to 72 hours.

Offer terms

Availability: Source to order
Unit price: Quoted per order
Minimum order: 1 unit
Ships from: Confirmed in your written offer
Dispatch: Lead time confirmed in your written offer
Incoterm: Agreed per order (Incoterms 2020)
Tax and duties: Quoted excluding UAE VAT, duties and import taxes unless stated
Payment: 30% deposit to secure the allocation, balance before dispatch; Bank transfer against pro forma invoice, or card payment link for approved orders
Condition: New, factory sealed

Overview

Each H200 SXM GPU provides 141 GB of HBM3e at 4.8 TB/s, 3,958 TFLOPS FP8 and 1,979 TFLOPS FP16 and BF16 (with sparsity). An eight-GPU server delivers over 32 PFLOPS of FP8 with about 1.1 TB of GPU memory.

HGX H200 runs the mature Hopper software stack and is often the fastest route to large-memory GPU capacity where Blackwell lead times are long. A liquid-cooled 8-GPU board (935-24287-2741-000) is also available.

Typical uses

LLM training and inference
HPC
Capacity expansion of existing Hopper clusters

Specifications

Platform
GPUs	8x NVIDIA H200 SXM
MIG	Up to 7 instances at 18 GB per GPU

Memory
GPU memory	141 GB HBM3e per GPU, about 1.1 TB total
Memory bandwidth	4.8 TB/s per GPU

Performance
FP8 Tensor	3,958 TFLOPS per GPU (with sparsity)
FP16 and BF16 Tensor	1,979 TFLOPS per GPU (with sparsity)
FP64 Tensor	67 TFLOPS per GPU

Interconnect
NVLink	900 GB/s per GPU
Host interface	PCIe Gen5, 128 GB/s

Power and thermal
GPU power	Up to 700 W per GPU, configurable
Cooling	Air-cooled (935-24287-2740-000) or liquid-cooled (935-24287-2741-000) baseboard

Source: manufacturer documentation, checked 16 October 2026. Tensor figures follow NVIDIA's stated basis; sparse figures are marked.

NVIDIA H200 page Lenovo Press LP1944

Edition and part numbers

Baseboard part numbers are taken from Lenovo Press LP1944. The 4-GPU HGX H200 board is 935-23087-2741-000.

Delivery and warranty

Delivery. Lead time confirmed in your written offer. Agreed per order (Incoterms 2020). Shipments are insured to declared value and tracked. Delivery to every destination is subject to export and sanctions screening; see export compliance.

Warranty. Stated per offer: manufacturer, OEM or supplier backed. Keep serial numbers with your invoice; our warranty and RMA process explains how claims are handled.

Cancellation. Written offers state their validity and cancellation terms. See returns and cancellations.

Questions about the HGX H200

Why choose HGX H200 over Blackwell?

Supply and compatibility. Hopper systems run the same software as existing H100 clusters and can be available sooner than Blackwell systems. For new reasoning workloads, Blackwell offers far higher FP4 throughput.

How quickly can you confirm availability?

Listed availability shows when supply was last verified. For other quantities or destinations, send a request and we confirm supply and price in writing.